An estimated 10 million children die yearly in low-income countries, the majority from infections easily prevented, diagnosed, and treated in wealthy countries. The actual sum and etiologic fractions are unknown, since few are diagnosed in life or classified by autopsy thereafter. Fever is often attributed erroneously to malaria in sub-Saharan Africa and to typhoid in South Asia. Limited cross-sectional data suggest that misdiagnosis is costly and deadly. Data to guide prevention and management of acute, undifferentiated fever in Sri Lanka are lacking. Our overarching hypothesis is that the development and rational use of new diagnostic tools will improve clinical outcomes and limit costs from acute febrile illness in resource-poor settings such as Sri Lanka. We posit that determination of the causal agents will inform public health efforts, improve clinical diagnosis, and reduce unnecessary morbidity, mortality, and costs by enabling the development of evidence-based clinical algorithms that include focused diagnostic testing. Therefore, we have the following specific aims: 1) To investigate the distribution and determinants of acute febrile illness in Sri Lanka, 2) To investigate new strategies for the diagnosis of acute febrile illness in Sri Lanka, and 3) To prospectively validate the prediction rules and to assess use and performance of the algorithm and related clinical outcomes and costs. The second aim will involve the validation of a multiplex real-time PCR assay for rickettsiae as a prototype for the rapid identification of difficult-to-culture emerging infections, derivation of a clinical prediction rule to assess etiology of acute fever in Sri Lanka, and use of decision analysis to evaluate strategies for managing acute febrile illness. The proposed 5-year career development period includes activities to provide the recipient with the additional training and skills required for her to conduct translational research and lead a multidisciplinary team as an independent investigator. The research will advance the field through integrated study of 1) traditional methods to define the epidemiology of acute febrile illness, 2) new molecular and analytic decision-making tools for rapid diagnosis of acute febrile illness and 3) feasibility and outcomes (health and costs) related to use of an algorithm to manage acute febrile illness in Sri Lanka. This work will provide data essential for prevention and management of acute febrile illness in Sri Lanka, a molecular diagnostic method for an emerging pathogen of worldwide importance (rickettsiae), and a general methodology for assessing approaches to improved diagnostic testing and related outcomes in resource-poor settings. . Globally many die with fever from infections that are never diagnosed. We will develop and study new tools to improve the diagnosis of infections characterized by fever, with the goal of reducing sickness and death from these infections and the cost of related medical care. The proposed 5-year training period will allow the applicant to gain skills and forge new collaborations to support a career in this neglected area.